Qwen2.5 72B Instruct vs Qwen3.8 27B
| Spec | Qwen2.5 72B Instruct | Qwen3.8 27B |
|---|---|---|
| Respan Index | 10.0 (#173) | 42.3 (#65) |
| Blended price per 1M | $0.37 | $0.154 |
| Context window | 131K | 262K |
| Input / output per 1M | $0.36 / $0.4 | $0.039 / $0.5 |
| Max output | 8K | - |
| Input | text | text, vision |
| Released | 2024-09-19 | 2026-08-14 |
Respan Index by category
Each category is scored 0 to 100, and how far ahead the winner is. Leads under a point are close to even.
| Category | Qwen2.5 72B Instruct | Qwen3.8 27B |
|---|---|---|
| Coding | 0.9 Qwen2.5 72B Instruct | 23.5 Qwen3.8 27B |
| Agents | 2.0 Qwen2.5 72B Instruct | 56.4 Qwen3.8 27B |
| Reasoning | 16.6 Qwen2.5 72B Instruct | 39.8 Qwen3.8 27B |
| Knowledge & language | 3.1* Qwen2.5 72B Instruct | 40.6 Qwen3.8 27B |
| Human preference | 32.2 Qwen2.5 72B Instruct | 55.6 Qwen3.8 27B |
* Estimated: no published scores in that category. How the Respan Index works
Benchmarks both report
Qwen3.8 27B leads on 2 of the 2 benchmarks both report. Higher is better. Each score links to where it was published.
| Benchmark | Qwen2.5 72B Instruct | Qwen3.8 27B |
|---|---|---|
| LMArena Elo | 1303 | 1435 |
| GPQA Diamond | 49.1% | 89.2% |
FAQ
Is Qwen2.5 72B Instruct better than Qwen3.8 27B?
Qwen3.8 27B scores 42.3 on the Respan Index (#65), ahead of Qwen2.5 72B Instruct at 10.0 (#173). Qwen3.8 27B leads on 2 of the 2 benchmarks both report.
Which is cheaper, Qwen2.5 72B Instruct or Qwen3.8 27B?
Qwen2.5 72B Instruct lists at $0.36 input and $0.4 output per 1M tokens; Qwen3.8 27B at $0.039 and $0.5. Qwen3.8 27B is 2.4x cheaper per token at list price.
Which has the larger context window?
Qwen2.5 72B Instruct accepts 131K tokens and Qwen3.8 27B 262K tokens.